HOW FAST DOES CLIMATE CHANGE?

The isotopic compositions of Quaternary ice from Greenland cores have been used extensively as proxies for regional and global climatic conditions. Preliminary statistical analysis suggests that the rates of change of isotopic composition (‰ / y) from consecutive annual samples can be modeled by a Poisson random model of compositional change. As such, it has been possible to differentiate, on the basis of a quantitative model, abrupt changes in isotopic composition from those associated with more typical (or background) variation.

Four subsets of data from the GRIP2 and GISP cores have been selected due to their differences in temporal resolution and their significance to the notion of abrupt climate change. For the Holocene records less than 1% of the observed rates were faster than predicted by a Poisson model; while for the late Pleistocene, approximately 5% of the changes were faster than predicted.

Despite the success of this approach, other less restrictive statistical models are worthy of consideration. For those portions of the ice-core isotopic records that display long-term stationarity, it is possible to model rate of change as a Gaussian distribution of deviations from the sample mean. The statistical validity and theoretical significance of these two approaches will be evaluated.

In addition, to ice core isotopic records, there exist a variety of other high-resolution proxies of climate change. Tree rings and lacustrine varves provide a useful alternative to the large-scale regional and global marine records of past climate. As such, it is valuable to evaluate the statistical characteristics of the rate of change of the continental proxies and contrasts their statistical characteristics with those exhibited by the ice core data.